End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.88,
-    "eval_accuracy": 0.8910891089108911,
-    "eval_loss": 0.2589154839515686,
-    "eval_runtime": 119.6967,
-    "eval_samples_per_second": 2.531,
-    "eval_steps_per_second": 0.084,
-    "total_flos": 3.607403494903972e+17,
-    "train_loss": 0.43609671706245057,
-    "train_runtime": 11592.208,
-    "train_samples_per_second": 1.175,
-    "train_steps_per_second": 0.009
+    "eval_accuracy": 0.9075907590759076,
+    "eval_loss": 0.24336053431034088,
+    "eval_runtime": 70.093,
+    "eval_samples_per_second": 4.323,
+    "eval_steps_per_second": 0.143,
+    "total_flos": 3.313083915385897e+17,
+    "train_loss": 0.4951313120978219,
+    "train_runtime": 9028.2693,
+    "train_samples_per_second": 1.509,
+    "train_steps_per_second": 0.012
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.88,
-    "eval_accuracy": 0.8910891089108911,
-    "eval_loss": 0.2589154839515686,
-    "eval_runtime": 119.6967,
-    "eval_samples_per_second": 2.531,
-    "eval_steps_per_second": 0.084
+    "eval_accuracy": 0.9075907590759076,
+    "eval_loss": 0.24336053431034088,
+    "eval_runtime": 70.093,
+    "eval_samples_per_second": 4.323,
+    "eval_steps_per_second": 0.143
 }

runs/Apr16_08-26-34_1569cada5f73/events.out.tfevents.1713265387.1569cada5f73.170.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ccd556462c9e3dd70b8923ae6c2b69a8a7a68de024182f287a52601e3761fa3
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.88,
-    "total_flos": 3.607403494903972e+17,
-    "train_loss": 0.43609671706245057,
-    "train_runtime": 11592.208,
-    "train_samples_per_second": 1.175,
-    "train_steps_per_second": 0.009
+    "total_flos": 3.313083915385897e+17,
+    "train_loss": 0.4951313120978219,
+    "train_runtime": 9028.2693,
+    "train_samples_per_second": 1.509,
+    "train_steps_per_second": 0.012
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8910891089108911,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-lungs-disease/checkpoint-86",
+  "best_metric": 0.9075907590759076,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-lungs-disease/checkpoint-105",
   "epoch": 4.883720930232558,
   "eval_steps": 500,
   "global_step": 105,
@@ -10,127 +10,127 @@
   "log_history": [
     {
       "epoch": 0.47,
-      "grad_norm": 14.870257377624512,
+      "grad_norm": 6.489995002746582,
       "learning_rate": 4.545454545454546e-05,
-      "loss": 1.1236,
+      "loss": 1.4339,
       "step": 10
     },
     {
       "epoch": 0.93,
-      "grad_norm": 20.39113998413086,
+      "grad_norm": 6.888110637664795,
       "learning_rate": 4.5212765957446815e-05,
-      "loss": 0.7104,
+      "loss": 0.814,
       "step": 20
     },
     {
       "epoch": 0.98,
-      "eval_accuracy": 0.7854785478547854,
-      "eval_loss": 0.5334098935127258,
-      "eval_runtime": 84.0412,
-      "eval_samples_per_second": 3.605,
-      "eval_steps_per_second": 0.119,
+      "eval_accuracy": 0.7491749174917491,
+      "eval_loss": 0.5313370227813721,
+      "eval_runtime": 66.4067,
+      "eval_samples_per_second": 4.563,
+      "eval_steps_per_second": 0.151,
       "step": 21
     },
     {
       "epoch": 1.4,
-      "grad_norm": 11.177618980407715,
+      "grad_norm": 15.805723190307617,
       "learning_rate": 3.9893617021276594e-05,
-      "loss": 0.4852,
+      "loss": 0.5569,
       "step": 30
     },
     {
       "epoch": 1.86,
-      "grad_norm": 10.540087699890137,
+      "grad_norm": 13.49924087524414,
       "learning_rate": 3.4574468085106386e-05,
-      "loss": 0.4067,
+      "loss": 0.4444,
       "step": 40
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.834983498349835,
-      "eval_loss": 0.42113950848579407,
-      "eval_runtime": 81.1616,
-      "eval_samples_per_second": 3.733,
-      "eval_steps_per_second": 0.123,
+      "eval_accuracy": 0.8910891089108911,
+      "eval_loss": 0.3199963867664337,
+      "eval_runtime": 66.8668,
+      "eval_samples_per_second": 4.531,
+      "eval_steps_per_second": 0.15,
       "step": 43
     },
     {
       "epoch": 2.33,
-      "grad_norm": 9.996329307556152,
+      "grad_norm": 8.2146577835083,
       "learning_rate": 2.925531914893617e-05,
-      "loss": 0.3041,
+      "loss": 0.3767,
       "step": 50
     },
     {
       "epoch": 2.79,
-      "grad_norm": 10.684453010559082,
+      "grad_norm": 12.676996231079102,
       "learning_rate": 2.393617021276596e-05,
-      "loss": 0.3307,
+      "loss": 0.3322,
       "step": 60
     },
     {
       "epoch": 2.98,
-      "eval_accuracy": 0.8778877887788779,
-      "eval_loss": 0.3033137917518616,
-      "eval_runtime": 82.0961,
-      "eval_samples_per_second": 3.691,
-      "eval_steps_per_second": 0.122,
+      "eval_accuracy": 0.8910891089108911,
+      "eval_loss": 0.31477591395378113,
+      "eval_runtime": 66.222,
+      "eval_samples_per_second": 4.576,
+      "eval_steps_per_second": 0.151,
       "step": 64
     },
     {
       "epoch": 3.26,
-      "grad_norm": 10.821243286132812,
+      "grad_norm": 9.924786567687988,
       "learning_rate": 1.8617021276595745e-05,
-      "loss": 0.261,
+      "loss": 0.2842,
       "step": 70
     },
     {
       "epoch": 3.72,
-      "grad_norm": 6.7758073806762695,
+      "grad_norm": 8.528675079345703,
       "learning_rate": 1.3297872340425532e-05,
-      "loss": 0.287,
+      "loss": 0.2975,
       "step": 80
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8910891089108911,
-      "eval_loss": 0.2589154839515686,
-      "eval_runtime": 80.5026,
-      "eval_samples_per_second": 3.764,
-      "eval_steps_per_second": 0.124,
+      "eval_accuracy": 0.8976897689768977,
+      "eval_loss": 0.28364694118499756,
+      "eval_runtime": 66.122,
+      "eval_samples_per_second": 4.582,
+      "eval_steps_per_second": 0.151,
       "step": 86
     },
     {
       "epoch": 4.19,
-      "grad_norm": 13.222726821899414,
+      "grad_norm": 16.571998596191406,
       "learning_rate": 7.97872340425532e-06,
-      "loss": 0.2682,
+      "loss": 0.2764,
       "step": 90
     },
     {
       "epoch": 4.65,
-      "grad_norm": 11.840774536132812,
+      "grad_norm": 6.849820613861084,
       "learning_rate": 2.6595744680851065e-06,
-      "loss": 0.2694,
+      "loss": 0.254,
       "step": 100
     },
     {
       "epoch": 4.88,
-      "eval_accuracy": 0.8844884488448845,
-      "eval_loss": 0.28530430793762207,
-      "eval_runtime": 79.9576,
-      "eval_samples_per_second": 3.79,
-      "eval_steps_per_second": 0.125,
+      "eval_accuracy": 0.9075907590759076,
+      "eval_loss": 0.24336053431034088,
+      "eval_runtime": 66.2106,
+      "eval_samples_per_second": 4.576,
+      "eval_steps_per_second": 0.151,
       "step": 105
     },
     {
       "epoch": 4.88,
       "step": 105,
-      "total_flos": 3.607403494903972e+17,
-      "train_loss": 0.43609671706245057,
-      "train_runtime": 11592.208,
-      "train_samples_per_second": 1.175,
-      "train_steps_per_second": 0.009
+      "total_flos": 3.313083915385897e+17,
+      "train_loss": 0.4951313120978219,
+      "train_runtime": 9028.2693,
+      "train_samples_per_second": 1.509,
+      "train_steps_per_second": 0.012
     }
   ],
   "logging_steps": 10,
@@ -138,7 +138,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 3.607403494903972e+17,
+  "total_flos": 3.313083915385897e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null